Spectrum sharing: Another way to better use the airwaves
Article By : Martin Rowe
The sharing of the 3.5-GHz band, which had been reserved for military radar, make another slice of spectrum available for wireless mobile broadband services. Before deployment can begin, Citizens Broadband Service Devices must be tested and certified.
When AM radio was in its infancy, stations in the U.S. broadcasted on one of three frequencies, which resulted in interference if two in the same area used the same frequency at the same time. Not until the Radio Act of 1927 were the airwaves regulated. Even as more wireless communications came online, there was enough spectrum to go around. Not anymore.
Today, spectrum is scarce and while the FCC has reallocated spectrum to accommodate services such as mobile broadband, it’s still not enough to handle the load. Some frequency bands have been reserved for the military, namely the desirable 3.5 GHz band (3550 MHz to 3700 MHz). Because we now have digital communications, it’s possible for multiple users to time-share frequencies, giving some priority over others. In 2015, the FCC changed its rules for this 150 MHz band, creating the Citizens Broadband Radio Service (CBRS), which allows licensed and unlicensed use in a spectrum slice once reserved for military radar. The opening of this band gives mobile operators more spectrum for lucrative mobile services. As you might expect, military use gets priority, something that wasn’t possible before today’s digital protocols.
Governed by FCC Part 96, CBRS consists of 15 10-MHz channels that use 3GPP LTE protocols for communication. Mobile operators using the CBRS band can aggregate channels for higher throughput.
“This will be the first time that commercial broadband users share spectrum dynamically with government users, and if it works, the FCC may allocate other currently protected RF bands for shared use,” said Michael Souryal, lead for the spectrum sharing support project within NIST’s Communications Technology Laboratory, in a press release. NIST’s role in CBRS development includes testing the overall system prior to deployment.
The key words here are “if it works.” That’s where testing will need to prove CBRS functionality.
CBRS can operate along the coasts or inland. It’s often a coastal issue because the Navy uses this frequency band for radar. Figure 1 shows a NIST simulation of wireless users from Maine to Cape Cod, Mass. Blue markers represent users that can continue operating in the affected channel. The others must vacate the channel, possibly relocating to another channel, when a higher-priority user needs the channel.
Figure 1 The coastline from Maine to Cape Cod has many potential users of the 3500 to 3700 MHz frequency band, with most centered around Boston. Source: NIST.
The CBRS system that controls access to the channels is called Spectrum Access System (SAS). An environmental sensing capability (ESC) system will alert the SAS when a higher priority user requests use of the channel. SAS selects which channels to allocate to users and their power levels.
CBRS incorporates a three-tiered system of granting channel priority (Figure 2), with the military holding the highest priority. Of the 150 MHz in the CBRS band, at least 80 MHz are available for general authorized access (GAA) use. Each device, called a Citizens Broadband Service Device (CBSD), must be registered in a database so that the cloud-based SAS can manage channel access.
Figure 2 A cloud-based Spectrum Access System consults a database of CBRS users to manage a three-tiered system for granting channel access.
The non-military/satellite devices are divided into two groups, designated Class A and Class B. Class A devices will generally be indoor devices such as small cells with a maximum of 1 W of output power. Class A devices can be installed by users, their associates, or employees. Class B devices will be base stations with up to 50 W of output power and must be installed by professionals such as those from wireless carriers. All devices must register their presence on the SAS database. Some Class A devices, such as those that can’t adequately identify their locations according to FCC Part 96 rules, will need to be professionally installed.
Before CBRS can become a reality, the sharing technology must be tested.
Testing
Testing of CBSDs will be performed by certified test labs using SAS software developed by members the Wireless Innovation Forum. Test software is called a test harness. According to Souryal, the Windows-based test harness, sunning on a local PC or on a server, will test CBSDs as they operate in a real-world scenarios; they won’t run in a test mode. Upon completing a test, a certified lab must submit test reports generated by the test harness to the FCC for a final check to verify Part 96 compliance before the device can be placed into service.
CBSD tests will verify RF frequency and signal power, plus the digital requirements of handshaking, protocols, and SAS interaction. The “RF test equipment” block in Figure 3 does not cover a DUT’s RF signal spectral characteristics such as mask tests, which will be tested separately per normal FCC procedures. Thus, it doesn’t provide calibrated measurements. Tests verify that a DUT’s RF signal turns on and off as needed.
Figure 3 The test setup for CBSDs emulates DUT operation in a real-world situation.
The CBSD Under Test block in Figure 3 refers to a domain proxy, which is for cases where a user—usually a wireless carrier—wants a single software manager for more than one CBSD. Either a CBSD or a domain proxy can communicate with the SAS. The test case for a domain proxy uses two CBSDs.
Test case data are stored in csv files, which will step through a series of JavaScript Object Notation (json) files. XML files will be lab specific for adding serial numbers, certificates, etc. Labs can edit for environment or specific DUT. The test harness will generate a test report with pass/fail summary.
SAS testing
Members of the Wireless Innovation Forum, including NIST, have developed a software harness for testing the SAS system itself, which runs in the cloud. A test harness (Figure 4) running on a PC or server emulates CBSDs and “peer” SAS systems, for more than one SAS can operate in a given area at any time. Most of the SAS testing involves verifying that protocols operate correctly—testing for proper registration of the CBSDs and correct passing of data. Test procedures also verify how a SAS calculates spectrum allocations to avoid interference. Based on calculations, the SAS can deny a user access to a channel by shutting it down or moving it to another channel. The test harness will also emulate RF signal strength using propagation models.
Figure 4 A local PC or server runs test harnesses that verify SAS functionality.
Details for SAS testing are publicly available in the document Test and Certification for Citizens Broadband Radio Service CBRS); Conformance and Performance Test Technical Specification; SAS as Unit Under Test (UUT) specification.
The following video from Wireless Innovation Forum (2 hrs.) covers the details of the technology: rules, protocol, security, registration process, and test cases for CBRS. Other videos are available on the Wireless Innovation Forum’s YouTube Channel.
—Martin Rowe covers test and measurement for EDN and EE Times. Contact him at martin.rowe@AspenCore.com
Related articles:
- Testing carrier aggregation in LTE-Advanced network infrastructure
- Alliance Seeks to Boost Indoor LTE
- How to test modern radars
- Phase noise and its changing role in radar design and test
- Radar gives cars eyes